One of the difficulties associated with keyword searching is providing context sensitive results. A search query based on several keywords can often result in thousands or even millions of potentially matching documents. Ideally, adding context to a keyword search can lead to more highly relevant documents being displayed closer to the top of the list of matching documents.
One type of context that can be used in keyword searching is location based context. Some search queries can explicitly identify a locality, such as “Seattle weather”. However, location based context can also be provided implicitly, such as by detecting and/or inferring a location for a user when a search query is potentially sensitive to location. Using such implicit location information, a query for “weather” can still generate results for “Seattle weather” based on information such as the location of the user's internet access or information from a user profile.